Adherence to the Mediterranean diet and lymphoma risk in the European Prospective Investigation into Cancer and Nutrition
Marta Solans1,2,3,, Yolanda Benavente1,4, Marc Saez1,2, Antonio Agudo5, Sabine Naudin6,Fatemeh Saberi
Hosnijeh7,8, Hwayoung Noh6, Heinz Freisling6, Pietro Ferrari6, Caroline Besson9,10,11, Yahya
Mahamat-Saleh10,11, Marie-Christine Boutron-Ruault10,11, Tilman Kühn12, Rudolf Kaaks12, Heiner Boeing13, Cristina
Lasheras14, Miguel Rodríguez-Barranco1, 15, Pilar Amiano1,16, Jose Maria Huerta1, 17, Aurelio
Barricarte1,18,19, Julie A Schmidt20, Paolo Vineis21, Elio Riboli22, Antonia Trichopoulou23,24, Christina Bamia 23,24, Eleni Peppa 23,24, Giovanna Masala25, Claudia Agnoli26, Rosario Tumino27, Carlotta Sacerdote28,
Salvatore Panico29, Guri Skeie30, Elisabete Weiderpass30,31,32,33, Mats Jerkeman34, Ulrika Ericson34,
Florentin Späth35, Lena Maria Nilsson36, Christina C Dahm37, Kim Overvad37, Anne Katrine Bolvig38, Anne
Tjønneland38,39,Silvia de Sanjose1,4,40, Genevieve Buckland41, Roel Vermeulen7,Alexandra Nieters42,
Delphine Casabonne1,4
1. Centro de Investigación Biomédica en Red: Epidemiología y Salud Pública (CIBERESP), Madrid, Spain
2. Research Group on Statistics, Econometrics and Health (GRECS), University of Girona, Girona, Spain
3. Epidemiology Unit and Girona Cancer Registry, Catalan Institute of Oncology, Girona, Spain 4. Unit of molecular and genetic epidemiology in infections and cancer, Catalan Institute of
Oncology (ICO-IDIBELL), Barcelona, Spain
5. Unit of Nutrition and Cancer. Cancer Epidemiology Research Program, Catalan Institute of Oncology (ICO-IDIBELL), Barcelona, Spain
6. Nutritional Methodology and Biostatistics Group, International Agency for Research on Cancer (IARC-WHO), Lyon, France
7. Institute for Risk Assessment Sciences, Division of Environmental Epidemiology, Utrecht University, Utrecht, The Netherlands
8. Immunology Department, Erasmus University Medical Center, Rotterdam, The Netherlands 9. Department of Hematology and Oncology, Hospital of Versailles, Le Chesnay, France
10. CESP, Fac. de médecine - Univ. Paris-Sud, Fac. de médecine - UVSQ, INSERM, Université Paris-Saclay, 94805, Villejuif, France
11. Gustave Roussy, F-94805, Villejuif, France
12. Division of Cancer Epidemiology, German Cancer Research Center (DKFZ), Heidelberg, Germany
13. Department of Epidemiology, German Institute of Human Nutrition (DIfE) Postdam-Rehbrücke, Nuthetal, Germany
14. Department of Functional Biology, Oviedo University, Spain
15. Escuela Andaluza de Salud Pública. Instituto de Investigación Biosanitaria ibs.GRANADA. Granada, Spain
16. Public Health Division of Gipuzkoa, Regional Government of the Basque Country, Donostia, Spain
17. Department of Epidemiology, Murcia Regional Health Council, IMIB-Arrixaca, Murcia, Spain 18. Navarra Public Health Institute, Pamplona, Spain
19. IdiSNA, Navarra Institute for Health Research, Pamplona, Spain
20. Cancer Epidemiology Unit, Nuffield Department of Population Health, University of Oxford, Oxford, United Kingdom.
21. Department of Epidemiology and Biostatistics, MRC-HPA Centre for Environment and Health, School of Public Health, Imperial College London, London, United Kingdom
22. School of Public Health, Imperial College London, London, United Kingdom 23. Hellenic Health Foundation, Athens, Greece
24. WHO Collaborating Center for Nutrition and Health, Unit of Nutritional Epidemiology and Nutrition in Public Health, Dept. of Hygiene, Epidemiology and Medical Statistics, School of Medicine, National and Kapodistrian University of Athens, Greece
25. Cancer Risk Factors and Life-Style Epidemiology Unit. Institute for Cancer Research, Prevention and Clinical Network - ISPRO, 50141 Florence, Italy
26. Epidemiology and Prevention Unit, Fondazione IRCCS Istituto Nazionale dei Tumori, Milano, Italy
27. Cancer Registry and Histopathology Department, "Civic - M. P. Arezzo" Hospital, ASP Ragusa, Italy
28. Unit of Cancer Epidemiology, Città della Salute e della Scienza University-Hospital and Center for Cancer Prevention (CPO), Turin, Italy
29. Dipartimento di Medicina Clinica e Chirurgia, Federico II University, Naples, Italy
30. Department of Community Medicine, University of Tromsø , The Arctic University of Norway, Tromsø, Norway
31. Department of Research, Cancer Registry of Norway - Institute of Population-Based Cancer Research, Oslo, Norway
32. Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden 33. Genetic Epidemiology Group, Folkhälsan Research Center, and Faculty of Medicine, Helsinki
University, Helsinki, Finland
34. Department of Clinical Sciences in Malmö, Lund University, Malmö, Sweden 35. Department of Radiation Sciences, Oncology, Umeå University, Umeå, Sweden
36. Public health and clinical Medicine, Nutritional Research and Arctic Research Centre, Umeå University, Umeå, Sweden.
37. Section for Epidemiology, Department of Public Health, Aarhus University, Aarhus, Denmark 38. Danish Cancer Society Research Center, Copenhagen, Denmark
40. PATH, Reproductive Health, Seattle, United States
41. NIHR Bristol Biomedical Research Centre Nutrition Theme, Bristol, United Kingdom
42. Center for Chronic Immunodeficiency (CCI), Medical Center - University of Freiburg, Faculty of Medicine, University of Freiburg, Freiburg, Germany
Corresponding author: Delphine Casabonne, Unit of Infections and Cancer. Cancer Epidemiology Research Programme. IDIBELL. Catalan Institute of Oncology. Av. Gran Via de l'Hospitalet 199-203 08908 L'Hospitalet de Llobregat, Spain Barcelona, Spain. E-mail: [email protected] Running title: Mediterranean diet and lymphoma in EPIC
Key words: lymphoma, Mediterranean diet, Europe, prospective studies, risk.
Abbreviations: arMED, adapted relative Mediterranean diet; BMI, body mass index; CLL/SLL, chronic lymphocytic leukemia/small lymphocytic leukemia; CI, confidence interval; DLBCL, diffuse large B-cell lymphoma; EPIC, European Prospective Investigation into Cancer and Nutrition; FL, follicular lymphoma; HL, Hodgkin lymphoma; HR, hazard ratio; MD, Mediterranean diet; MM/PCN, multiple myeloma/plasma cell neoplasm; NHL, Non-Hodgkin lymphoma; WCRF/AICR World Cancer Research Fund/American Institute for Cancer Research.
Novelty and impact: Known risk factors explain only a small proportion of lymphoma cases. Several studies have pointed out the potential role of dietary factors on lymphoma risk, but evidence is still inconclusive. Here, using data from the European Prospective Investigation into Cancer and Nutrition study, the authors found for the first time that adherence to a Mediterranean diet was modestly associated with a reduced risk of overall lymphoma. Further studies are needed to confirm these findings.
Funding: European Commission (DG-SANCO), International Agency for Research on Cancer, Danish Cancer Society (Denmark), Ligue Contre le Cancer, Institut Gustave Roussy, Mutuelle Generale de l’Education Nationale, Institut National de la Sante et de la Recherche Medicale (INSERM) (France), German Cancer Aid, German Cancer Research Center (DKFZ), Federal Ministry of Education and Research (BMBF), Deutsche Krebshilfe, Deutsches Krebsforschungszentrum and Federal Ministry of Education and Research (Germany), the Hellenic Health Foundation (Greece), Associazione Italiana per la Ricerca sul Cancro-AIRC-Italy and National Research Council (Italy), Dutch Ministry of Public Health, Welfare and Sports (VWS), Netherlands Cancer Registry (NKR), LK Research Funds, Dutch Prevention Funds, Dutch ZON (Zorg Onderzoek Nederland), World Cancer Research Fund (WCRF), Statistics Netherlands (The Netherlands); Grant number: ERC2009-AdG 232997; Grant sponsors: Nordforsk, Nordic Centre of Excellence programme on Food, Nutrition and Health (Norway); German Federal Ministry of Education and Research (BMBF 01EO1303); Grant sponsor: Spanish Ministry of Economy and
Competitiveness - Carlos III Institute of Health cofunded by FEDER funds/European Regional Develpment Fund (ERDF) - a way to build Europe; Grant numbers: PI13/00061 (to Granada), PI13/01162 (to EPIC-Murcia, Regional Governments of Andalucıa, Asturias, Basque Country, Murcia and Navarra), PI17/01280 and PI14/01219 (to Barcelona); Centro de Investigación Biomédica en Red: Epidemiología y Salud Pública (CIBERESP, Spain)]; Grant sponsor: Agència de Gestió d'Ajuts Universitaris i de Recerca (AGAUR); Grant number (2017SGR1085); Grant sponsor: ISCIII RETIC; Grant number: RD06/0020 (Spain); Grant sponsors: Swedish Cancer Society, Swedish Research Council and County Councils of Skåne and Vasterbotten (Sweden) and Cancer Research UK; Grant numbers: 14136 (to EPIC-Norfolk), C570/A16491 and C8221/A19170 (to EPIC-Oxford); Grant sponsor: Medical Research Council; Grant numbers: 1000143 (to EPIC-Norfolk), MR/M012190/1 (to EPIC-Oxford, UK).
Availability of data and materials: For information on how to submit an application for gaining access to EPIC data and/or biospecimens, please follow the instructions at http://epic.iarc.fr/access/index.php. Conflicts of interest: none
Article category: research article Word count:
Abstract: 250 words Text: 3,417 words
Tables and figures: 3 tables
Abstract
There is growing evidence of the protective role of the Mediterranean diet (MD) on cancer. However, no prospective study has yet investigated its influence on lymphoma. We evaluated the association between adherence to the MD and risk of lymphoma and its subtypes in the European Prospective Investigation into Cancer and Nutrition (EPIC) study. The analysis included 476,160 participants, recruited from ten European countries between 1991 and 2001. Adherence to the MD was estimated through an adapted relative Mediterranean diet (arMED) score excluding alcohol.Cox proportional hazards regression models were used while adjusting for potential confounders. During an average follow-up of 13.9 years, 3,136 lymphomas (135 Hodgkin lymphoma (HL), 2,606 non-Hodgkin lymphoma and 395 lymphoma NOS) were identified. Overall, a 1-unit increase in the arMED score was associated with a 2% lower risk of lymphoma (95% CI: 0.97; 1.00, p-trend=0.03) while a statistically non-significant inverse association between a high versus low arMED score and risk of lymphoma was observed (HR: 0.91 (95% CI 0.80; 1.03), p-trend=0.12). Analyses by lymphoma subtype did not reveal any statistically significant associations. Albeit with small numbers of cases (N= 135), a suggestive inverse association was found for HL (HR 1-unit increase= 0.93 (95% CI: 0.86; 1.01), p-trend=0.07). However, the study may have lacked statistical power to detect small effect sizes for lymphoma subtype. Our findings suggest that an increasing arMED score was inversely related to the risk of overall lymphoma in EPIC, but not by subtypes. Further large prospective studies are warranted to confirm these findings.
1. Introduction
Lymphomas are a heterogeneous group of malignancies particularly prevalent in Western countries. Although some lymphoid neoplasms have been consistently linked to certain infections and severe immunosuppression, their etiology remains elusive, and evidence from epidemiologic studies increasingly points to etiologic heterogeneity among subtypes1.
An increase in the incidence of lymphoma has been observed in many regions during the last decades2, and a change in lifestyle might be one possible explanation for this pattern. However,
there is limited evidence regarding extrinsic-risk factors, particularly diet, and lymphoma risk3,4. In
the recently released Third Expert Report on ‘Diet, Nutrition, Physical Activity, and Cancer’ by the World Cancer Research Fund/American Institute for Cancer Research (WCRF/AICR)5, no
conclusive associations for specific dietary factors and hematological malignancies were reported. Studies from the European Prospective Investigation into Nutrition and Cancer (EPIC) study have neither shown consistent associations between consumption of meat and dairy6 nor
vegetables7 and overall lymphoma risk, although some statistically significant associations were
seen for several lymphoma subtypes.
There is evidence that the Mediterranean diet (MD) has a protective role on risk of overall8 and
specific types of cancer9, such as breast10, colorectal11 or gastric cancers12. However,
epidemiological research into the effect of a MD pattern on lymphoma remains limited. To our knowledge, no study has yet evaluated the influence of validated a priori MD score on lymphoma risk, while studies on a posteriori healthy-like dietary patterns have yielded inconsistent results for overall lymphoma as well as its subtypes13–15.
The aim of this study is to investigate the association between adherence to the Mediterranean dietary pattern and lymphoma risk within the EPIC population. The EPIC study provides the opportunity to examine this relation in a prospective design and within a European population with a wide spectrum of dietary habits.
2. Material and Methods
EPIC is a large prospective cohort study designed to investigate the relationship between diet, lifestyle, environmental factors and cancer. The rational, full methods and study design have been described previously16,17. In brief, 521,324 subjects, mostly aged 30 to 70 years, were
recruited between 1992-2000 in 23 centers from ten European countries (Denmark, France, Germany, Greece, the Nederland’s, Italy, Norway, United Kingdom, Spain and Sweden). The ethical review boards from the International Agency for Research on Cancer (IARC) and all local participating centers approved the study, and all participants gave their informed consent.
Of the 521,324 EPIC cohort participants, we excluded prevalent cancer cases (n= 25,184), subjects with missing follow-up information (n= 4,148), with incomplete/ no dietary information (n= 6,259), or those in the highest and lowest 1% of the distribution for the ratio of energy intake to estimate energy requirement (n= 9,573). Therefore, the current analysis was based on 476,160 subjects among whom 3,136 incident lymphoma cases occurred.
Data collection
Validated country-specific questionnaires were used to record the usual diet during the previous year17,18; namely through quantitative or semi-quantitative food frequency questionnaires (FFQs)
(administered through a personal interview or self-administered), although few countries used semi-quantitative FFQs combined with a food record. Lifestyle questionnaires were used to obtain information on sociodemographic characteristics, physical activity, reproductive history, use of oral contraceptives and hormone replacement therapy, medical history and alcohol and tobacco consumption. Anthropometric measures were also ascertained at recruitment.
Exposure assessment – arMED
The level of adherence to the MD was assessed using the adapted relative MD (arMED) score10,
which excludes alcoholic beverages as they have been inversely associated with several lymphoma subtypes19. The scoring system, adapted from the original index designed by
Trichopoulou et al.20, has been detailed previously12. In brief, the arMED is a 16-point linear score
that incorporates eight key dietary components: six components presumed to reflect the MD [fruit (including nuts and seeds), vegetables, legumes, fish (including seafood), olive oil and cereals] and two components consumed in low quantity in the MD (dairy products and meat). Intake of each component was calculated as a function of energy density (g/day/1000kcal) and divided into tertiles (estimated using the overall study population). A score of 0 to 2 was assigned for the first, second and third tertile of intakes for the components presumed to fit the MD, while the scoring
was inverted for the components presumed to not fit the MD (giving a lower score for higher intakes). The scoring for olive oil was adapted owing the low consumption of non-Mediterranean countries, by assigning 0 to non-consumers, 1 for subjects below the median and 2 for subjects equal or above this median (the median was calculated using the overall study population and considering only consumers). The points were summed to define the arMED score, that ranged from 0 to 16 (from the lowest to the highest adherence), and was in turn divided into three categories: low (0-5), medium(6-9) and high(10-16) as described previously12.
Follow-up and outcome assessment
Incident lymphoma cancer cases were identified by population cancer registries for Denmark, Italy, the Netherlands, Norway, Spain, Sweden and the United Kingdom. A combination of methods was used in France, Germany and Greece, including cancer and pathology registries, health insurance records, and active follow-up contacting participants or their next-of-kin. Mortality data were also obtained from regional or national mortality registries. The follow-up period was defined from the age at recruitment to the age at first cancer diagnosis, death or last complete follow-up, depending on which occurred first. Censoring dates for the last complete follow-up ranged from June 2008 to December 2013, depending on the EPIC center.
Initially, the diagnosis of lymphoma cases was based on the second revision of the International Classification of Diseases for Oncology (O-2). Later, all cases were reclassified into the ICD-O-3, using a conversion program available on the web site of the Surveillance Epidemiology and End Results (SEER) program (http://seer.cancer.gov/tools/conversion/ICD02- 3manual.pdf) and involving a pathology expert and experts from the EPIC centers. Because not all ICD-O-2 diagnostics can be translated unequivocally into the current classification, we left the respective lymphomas unclassified (not otherwise specified ‘‘NOS’’) when further detailed specification failed. Finally, the InterLymph Pathology Working Group classification, which is based in the 2008 WHO classification, was used to categorize lymphoma histologic subtypes1.
In the current analysis, the following groups were considered: Hodgkin lymphoma (HL) and non-Hodgkin lymphoma (NHL); within NHL, mature B-cell lymphoma and mature T/NK-cell lymphoma; and among mature B-cell lymphoma, the following entities: diffuse large B-cell lymphoma (DLBCL), follicular lymphoma (FL), chronic lymphocytic leukemia (CLL) (including small lymphocytic leukemia), multiple myeloma/plasma cell neoplasm (MM/PCN), and other B-cell lymphoma (i.e. those cases in which the B-cell lymphoma subtype is unknown or does not fall
within the above mentioned subtypes). Other entities were not considered due to small numbers (Table 1).
Statistical analysis
Cox proportional hazard models were used to estimate the hazard ratio (HR) and 95% confidence intervals (CI) of the association between the arMED score and lymphoma risk. The arMED score was analyzed both as a continuous variable (per 1-unit increase) and as a categorical variable (low, medium and high level of adherence). In addition, Cox models were fitted with the arMED ordinal variable as continuous to test for linear trend for comparison with published literature on solid cancers. Two models with two levels of adjustment were used: a basic model, stratified by center, sex and age at recruitment (in 1-year categories) and a multivariable model, further adjusted for body mass index (BMI) (<25, 25-30, ≥30 kg/m2), total energy intake (continuous,
kcal/day), educational level (no formal education, primary school, secondary school, technical or professional training, University, unknown [3.6%]), height (continuous, cm), physical activity level based on the Cambridge physical activity index (inactive, moderately inactive, moderately active, active, unknown [1.9%]), smoking status (never, former, current and, unknown [2.0%]), and alcohol intake at recruitment (continuous, g/day). We tested for interaction by age, sex, alcohol intake and smoking by including a cross-product term along with the armed score (continuous) in the multivariable Cox model. The statistical significance of the cross-product term was evaluated using likelihood ratio test.
Sensitivity analyses were performed by repeating main Cox analyses (i) including alcohol in the score computation, ii) censoring participants and excluding cases with less than two years of follow-up (n=259 cases), (iii) excluding participants without complete data (n=226 cases), and (iv) restricting HL analysis to classical HL cases. Schoenfeld residuals were used to ensure that the proportional hazard assumption was met in all models. Two-sided p-values were reported with statistical significance set at p<0.05. All analyses were performed by using STATA statistical software, version 14 (Stata Corporation, College Station, Texas).
3. Results
During an average follow-up of 13.9 years, 3,136 lymphoma cases (2,606 NHL, 135 HL and 395 lymphoma NOS) were diagnosed. A detailed distribution of cases by lymphoma subtype and country is displayed in Table 1. As expected, the highest levels of the arMED score were found in Mediterranean regions (i.e. Greece, Spain and Italy), while the lowest scores were found in
Sweden, the Netherland’s and Denmark. Baseline characteristics of the study participants according to category of the arMED score are shown in Table 2. In general, participants with a higher arMED score were more likely to be women, slightly younger, never smokers, physically inactive, and to have a higher educational level and lower alcohol intake compared to those with a low arMED score.
Table 3 shows basic and multivariable HR estimates for category of arMED score associated with lymphoma risk, overall and by subgroups. Overall, a 1-unit increase in the arMED score was associated with a 2% lower risk of lymphoma (95% CI: 0.97; 1.00, p-trend=0.03) while a statistically non-significant inverse association between a high versus low arMED score and risk of lymphoma was observed (HR: 0.91 (95% CI 0.80; 1.03), p-trend=0.12). No statistically significant associations were observed between the arMED score and HL, NHL or any NHL subtypes. However, albeit with smaller numbers of cases, the lowest HR were observed for HL (HRhigh vs. low=0.64 (95% CI: 0.34; 1.19), p-trend=0.16; HR1-unit increase=0.93 (95% CI: 0.86; 1.01),
p-trend=0.07). Following restriction to classical HL (n=127), although the results were still not statistically significant, the inverse association seemed to be strengthened (HRhigh vs. low= 0.57
(95% CI: 0.30; 1.09), p-trend=0.09; HR1-unit increase=0.76 (95% CI: 0.55; 1.05), p-trend= 0.09) (data
not shown). The results for the unclassified lymphoma (N= 395) as well as following restriction to lymphoma with known subtype classification did not modify materially the association (HR for a 1-unit increase in the arMED score: 0.98 (95% CI: 0.94; 1.03), value= 0.43, and 0.98 (95% CI: 0.96 to 1.00), p-value= 0.04, respectively) (data not shown).
No significant modifications in the association between lymphoma or its subtypes and the arMED score were observed for age, sex, and alcohol intake and smoking (Supplementary material, Table S1). Similarly, no significant differences were observed among countries (Supplementary material, Figure S1).
In sensitivity analyses, including alcohol in the scoring did not affect the estimates for most of the lymphoma subtypes, although statistically significant associations were found for HL (HR1-unit increase=0.93 (95% CI: 0.86; 0.99), p-trend= 0.04)) and DLBCL (HR1-unit increase=0.96 (95% CI: 0.92;
0.99), p-trend= 0.02)) (Supplementary material, Table S2). Moreover, excluding the first 2 years of follow-up and those individuals with no information on adjustment variables from the analyses did not materially alter the results (data not shown).
4. Discussion
This is the first prospective study to investigate the association between a Mediterranean dietary score and risk of lymphoma and its subtypes. Our findings suggest that a higher adherence to the MD is modestly associated with a lower risk of lymphoma.
Current evidence on the adherence to MD and the etiology of lymphoma is scarce. To our knowledge, only one case-control study has evaluated the association of an a priori MD score and lymphoma, yielding no significant association for overall cases and only reporting an inverse association for DLBCL21. However, those results require cautious interpretation due to the
retrospective study design, small sample size (322 cases for the particular subgroup analysis), and the use of a non-validated adaptation of the arMED score which did not include olive oil, white meat or dairy products. Other studies have extracted healthy-like dietary patterns from their population using data-driven analyses13–15, but also failed in finding patterns with all the MD
features (e.g. none of them included olive oil). With the exception of the prospective study of Erber et al.14 who reported an inverse association between a pattern rich in vegetables and fruits
and NHL among Caucasian women, none of the other case-control studies found any associations with overall HL15 or NHL13. However, given the heterogeneity in types of foods eaten
within these patterns, the range and absolute amounts of food intakes and cut-offs used to define adherence, direct comparison of study results should be made with caution.
Similarly, there is a lack of consistency for associations between questionnaire-derived dietary components and lymphoma risk. In the 2007 report by the WCRF/AICR22, the panel did not make
any judgements regarding the causality of associations between specific dietary factors and lymphoid neoplasms. However, several suggestive associations were pointed out: i) vegetables, fruits, and alcoholic beverages were associated with decreased risk of lymphoma, ii) meat, total fat, and body fatness with increased risk of lymphoma, and iii) dairy products with increased risk of NHL. Although numerous studies have subsequently emerged, most of them targeted NHL patients, were mainly case-control studies and did not show consistent associations. Indeed, in the recently released third report of the WCRF/AICR, no additional information has been provided for hematological neoplasms5. Recent meta-analyses have shed light on this relationship: i) two
meta-analyses found inverse relationships of vegetables and vegetables and fruit (combined) intake with NHL23,24, ii) another on foods of animal origin (including red, processed and white
meat, fish and seafood, dairy products and eggs) reported positive associations with red meat and dairy intake and NHL25, iii) while cohort studies on specific micronutrients point to null
associations between supplemented vitamins A, C and E, total vitamin D intake, as well as dietary lycopene intake, and risk of NHL26. Overall, our results suggest that, more than specific
dietary components, is the combined effect of a range of nutrients along with the putative biological interactions that take place between them that may be mediating the modest influence of adherence to the MD on lymphoma risk
In an exploratory analysis, we observed strong inverse associations with HL albeit statistically non-significant. Given the small number of HL (N= 135), we may have lacked statistical power to detect significant associations within this subgroup. Previous studies of diet and HL risk are scarcer and limited in power6,7,15,21,27–34. Although no consistent associations have been reported
for single-food items, the first study on dietary patterns and HL provided some insight15. The
authors found a suggestive inverse association between a diet characterized by high intake of fruit and low-fat dairy products and mixed cellularity HL. In addition, positive associations were reported between Western-like patterns (rich in meat or desserts and sweets) and specific HL entities and age-groups. Together, these and our results suggest that HL might be a lymphoma prone to be influenced by dietary patterns. Thus, further studies with prospective design and with histological subtype-specific analyses, feasible though pooling data from consortium studies, are warranted.
The original MD score included alcohol scored dichotomous variable12: two points were assigned
for moderate consumers (5–25 g/day for women and 10–50 g/day for men) and 0 points for those above and below the sex-specific range, owing its beneficial effects if consumed in moderation. Convincing evidence suggests that alcohol increases the risk of several carcinomas (e.g. mouth, pharynx and larynx, esophagus, liver, colorectal, breast and stomach)5 and indeed, the
WCRF/AICR recommendations currently promote lowering alcohol consumption5. However,
accumulating evidence showed a moderate inverse association between increasing alcohol intake and NHL, especially on DLCBL and FL19. Thus, for the current analyses, alcohol was not
included in the score, and models were adjusted and further stratified by alcohol intake. In sensitivity analyses, including alcohol in the score, statistically significant associations were found for HL and DLBCL. The influence of alcohol in lymphomagenesis remains largely unknown, and further studies are needed to clarify its role and possible interaction with dietary factors.
Certain dietary features of the MD and their potentially anti-carcinogenic mechanisms make the association with lymphoma plausible from a mechanistic point of view. The abundance of plant-based foods in the MD provides a diet rich in flavonoids, carotenoids, vitamin C or E, whose
important antioxidant properties can neutralize free radicals or prevent DNA damage22,35. Indeed,
total antioxidant intake has been inversely associated with lymphoma36,37. Moreover, it presents
a high monounsaturated to saturated fatty acid ratio, and its believed that circulating fatty acids may influence lymphoma risk by modulating inflammation or lymphocyte membrane stability38. In
addition, several studies have consistently linked chronic inflammation or autoimmune conditions with lymphomagenesis3,39 and reported associations between plasma levels of cytokines, or other
inflammatory markers, and lymphoma40–42. Recent studies are supporting the inflammatory
potential of diet43; in particular for lymphoid neoplasms, positive associations have been recently
reported between a pro-inflammatory dietary score and NHL44. Thus, the MD’s favorable fatty
acid profile, as well as a high intake of fiber, vitamins and flavonoids, may be relevant owing its properties43.
Interestingly, for the years 2000-2002, the incidence of total lymphoid malignances, in particular HL, was higher in Southern Europe (Italy, Malta, and Spain) in comparison with other European regions45. We are not aware of studies that have attempted to correlate known risk factors for
lymphoid neoplasms with regional variations in incidence. Further research on other dietary patterns (e.g. adherence to a Western-like or a pro-inflammatory diet) or the nutrition transition towards non-Mediterranean dietary patterns in many Mediterranean countries46 is warranted to
elucidate these incidence patterns.
Limitations of our study should be considered when interpreting the results, including potential measurement errors derived from dietary questionnaires, which could lead to systematic and random errors when estimating adherence to the MD. Although our adjustment for total energy intake would partly remove some of these errors47,48 we cannot rule out that they have affected
risk estimates. In addition, we were unable to take into account any possible changes in dietary and lifestyle habits over time. In particular, cases might have modified their diet during the early prediagnostic period of the disease, although sensitivity analyses excluding incident cases diagnosed in the first 2 years of follow-up did not alter the association. Moreover, we lacked data on other potential confounders (e.g. occupational exposures or pesticide use) and, despite adjusting for all known lymphoma’s risk factors, residual confounding by other unmeasured or unknown exposure cannot be dismissed. In addition, given the number of comparisons performed, we cannot exclude chance findings. Moreover, despite the large number of enrolled subjects at baseline, the number of observed incident cases of some lymphoma subtype was low (e.g. 135 HL). Therefore, the study might not have sufficient power to detect significant associations within those subgroups. Finally, the arMED score has also limitations, as similar
weight is given to each component and the foods within them, assuming that they have the same effects on health. However, according to current evidence, these groupings did not include dietary components which have distinct effects on lymphoma risk. In addition, the EPIC study included participants from both Mediterranean and non-Mediterranean regions, which may have distinct dietary intakes, food sources and socio-demographic, anthropometric and lifestyle characteristics that cannot be considered when using cohort-wide tertiles to construct the score. However, several studies have found similar associations’ when using study-wide or country-specific cut-offs8,49 and no interactions were found by country in our study.
The strengths of this study include its prospective design, long follow-up and large sample size which allowed us to carry out analyses by lymphoma subentities. In addition, its multi-centric European design allowed the inclusion of a geographically diverse population, covering a wide range of dietary patterns and lifestyle habits. Finally, we assessed adherence to the MD with a widely-used score in cancer epidemiology, which directly includes olive oil intake, a key feature of this dietary pattern. Moreover, it uses tertiles of intake as cut-offs instead of the frequently used medians to give a better distribution of the subjects with different intakes.
In summary, this is the first prospective study to examine the association between the MD and lymphoma risk. Our findings suggest that an increasing arMED score, reflecting adherence to the MD was modestly inversely associated with the risk of overall lymphoma. Further studies are needed to confirm these findings.
Acknowledgments
We thank all participants of the EPIC study and Bertrand Hémon at IARC for his valuable work and technical support with the EPIC database.
References
1. Turner JJ, Morton LM, Linet MS, Clarke CA, Kadin ME, Vajdic CM, Monnereau A, Maynadie M, Chiu BC-H, Marcos-Gragera R, Costantini AS, Cerhan JR, et al. InterLymph hierarchical classification of lymphoid neoplasms for epidemiologic research based on the WHO classification (2008): update and future directions. Blood 2010;116:e90-8.
2. Fitzmaurice C, Allen C, Barber RM, Barregard L, Bhutta ZA, Brenner H, Dicker DJ, Chimed-Orchir O, Dandona R, Dandona L, Fleming T, Forouzanfar MH, et al. Global, Regional, and National Cancer Incidence, Mortality, Years of Life Lost, Years Lived With Disability, and Disability-Adjusted Life-years for 32 Cancer Groups, 1990 to 2015: A Systematic Analysis for the Global Burden of Disease Study. JAMA Oncol 2017;3:524–48. 3. Morton LM, Slager SL, Cerhan JR, Wang SS, Vajdic CM, Skibola CF, Bracci PM, de Sanjose S, Smedby KE, Chiu BCH, Zhang Y, Mbulaiteye SM, et al. Etiologic heterogeneity among non-Hodgkin lymphoma subtypes: the InterLymph Non-Hodgkin Lymphoma Subtypes Project. J Natl Cancer Inst Monogr 2014;2014:130–44.
4. Roman E, Smith AG. Epidemiology of lymphomas. Histopathology 2011;58:4–14.
5. World Cancer Research Fund/American Institute for Cancer Research. Diet, Nutrition, Physical Activity, and Cancer: a Global Perspective. Continuous Update Project Expert Report 2018. Available at dietandcancerreport.org.
6. Rohrmann S, Linseisen J, Jakobsen MU, Overvad K, Raaschou-Nielsen O, Tjonneland A, Boutron-Ruault MC, Kaaks R, Becker N, Bergmann M, Boeing H, Khaw K-T, et al. Consumption of meat and dairy and lymphoma risk in the European Prospective Investigation into Cancer and Nutrition. Int J cancer 2011;128:623–34.
7. Rohrmann S, Becker N, Linseisen J, Nieters A, Rudiger T, Raaschou-Nielsen O, Tjonneland A, Johnsen HE, Overvad K, Kaaks R, Bergmann MM, Boeing H, et al. Fruit and vegetable consumption and lymphoma risk in the European Prospective Investigation into Cancer and Nutrition (EPIC). Cancer Causes Control 2007;18:537–49.
8. Couto E, Boffetta P, Lagiou P, Ferrari P, Buckland G, Overvad K, Dahm CC, Tjonneland A, Olsen A, Clavel-Chapelon F, Boutron-Ruault M-C, Cottet V, et al. Mediterranean dietary pattern and cancer risk in the EPIC cohort. Br J Cancer 2011;104:1493–9.
9. Schwingshackl L, Schwedhelm C, Galbete C, Hoffmann G. Adherence to Mediterranean Diet and Risk of Cancer: An Updated Systematic Review and Meta-Analysis. Nutrients 2017;9.
10. Buckland G, Travier N, Cottet V, Gonzalez CA, Lujan-Barroso L, Agudo A, Trichopoulou A, Lagiou P, Trichopoulos D, Peeters PH, May A, Bueno-de-Mesquita HB, et al. Adherence to the mediterranean diet and risk of breast cancer in the European prospective investigation into cancer and nutrition cohort study. Int J cancer 2013;132:2918–27.
11. Bamia C, Lagiou P, Buckland G, Grioni S, Agnoli C, Taylor AJ, Dahm CC, Overvad K, Olsen A, Tjonneland A, Cottet V, Boutron-Ruault M-C, et al. Mediterranean diet and colorectal cancer risk: results from a European cohort. Eur J Epidemiol 2013;28:317–28. 12. Buckland G, Agudo A, Lujan L, Jakszyn P, Bueno-de-Mesquita HB, Palli D, Boeing H,
Carneiro F, Krogh V, Sacerdote C, Tumino R, Panico S, et al. Adherence to a Mediterranean diet and risk of gastric adenocarcinoma within the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort study. Am J Clin Nutr 2010;91:381– 90.
13. Ollberding NJ, Aschebrook-Kilfoy B, Caces DBD, Smith SM, Weisenburger DD, Chiu BC-H. Dietary patterns and the risk of non-Hodgkin lymphoma. Public Health Nutr 2014;17:1531–7.
14. Erber E, Maskarinec G, Gill JK, Park S-Y, Kolonel LN. Dietary patterns and the risk of non-Hodgkin lymphoma: the Multiethnic Cohort. Leuk Lymphoma 2009;50:1269–75. 15. Epstein MM, Chang ET, Zhang Y, Fung TT, Batista JL, Ambinder RF, Zheng T, Mueller
NE, Birmann BM. Dietary pattern and risk of hodgkin lymphoma in a population-based case-control study. Am J Epidemiol 2015;182:405–16.
16. Riboli E, Kaaks R. The EPIC Project: rationale and study design. European Prospective Investigation into Cancer and Nutrition. Int J Epidemiol 1997;26 Suppl 1:S6-14.
17. Riboli E, Hunt KJ, Slimani N, Ferrari P, Norat T, Fahey M, Charrondiere UR, Hemon B, Casagrande C, Vignat J, Overvad K, Tjonneland A, et al. European Prospective Investigation into Cancer and Nutrition (EPIC): study populations and data collection.
18. Margetts BM, Pietinen P. European Prospective Investigation into Cancer and Nutrition: validity studies on dietary assessment methods. Int. J. Epidemiol.1997;26 Suppl 1:S1-5. 19. Psaltopoulou T, Sergentanis TN, Ntanasis-Stathopoulos I, Tzanninis IG, Tsilimigras DI
DM. Alcohol consumption and risk of hematological malignancies: A meta-analysis of prospective studies. Int J cancer 2018;143:486–95.
20. Trichopoulou A, Costacou T, Bamia C, Trichopoulos D. Adherence to a Mediterranean diet and survival in a Greek population. N Engl J Med 2003;348:2599–608.
21. Campagna M, Cocco P, Zucca M, Angelucci E, Gabbas A, Latte GC, Uras A, Rais M, Sanna S, Ennas MG. Risk of lymphoma subtypes and dietary habits in a Mediterranean area. Cancer Epidemiol 2015;39:1093–8.
22. World Cancer Research Fund / American Institute for Cancer Research. Food, Nutrition, Physical Activity, and the Prevention of Cancer: a Global Perspective. Washington DC: AICR, 2007.
23. Chen G-C, Lv D-B, Pang Z, Liu Q-F. Fruits and vegetables consumption and risk of non-Hodgkin’s lymphoma: a meta-analysis of observational studies. Int J cancer 2013;133:190–200.
24. Sergentanis TN, Psaltopoulou T, Ntanasis-Stathopoulos I, Liaskas A, Tzanninis I-G, Dimopoulos M-A. Consumption of fruits, vegetables, and risk of hematological malignancies: a systematic review and meta-analysis of prospective studies. Leuk
Lymphoma 2018;59:434–47.
25. Caini S, Masala G, Gnagnarella P, Ermini I, Russell-Edu W, Palli D, Gandini S. Food of animal origin and risk of non-Hodgkin lymphoma and multiple myeloma: A review of the literature and meta-analysis. Crit Rev Oncol Hematol 2016;100:16–24.
26. Psaltopoulou T, Ntanasis-Stathopoulos I, Tsilimigras DI, Tzanninis I-G, Gavriatopoulou M, Sergentanis TN. Micronutrient Intake and Risk of Hematological Malignancies in Adults: A Systematic Review and Meta-analysis of Cohort Studies. Nutr Cancer 2018;1–19.
27. Tavani A, Pregnolato A, Negri E, Franceschi S, Serraino D, Carbone A, La Vecchia C. Diet and risk of lymphoid neoplasms and soft tissue sarcomas. Nutr Cancer 1997;27:256– 60.
28. Middleton B, Byers T, Marshall J, Graham S. Dietary vitamin A and cancer--a multisite case-control study. Nutr Cancer 1986;8:107–16.
29. Gao Y, Li Q, Bassig BA, Chang ET, Dai M, Qin Q, Zhang Y, Zheng T. Subtype of dietary fat in relation to risk of Hodgkin lymphoma: a population-based case-control study in Connecticut and Massachusetts. Cancer Causes Control 2013;24:485–94.
30. Chang ET, Canchola AJ, Clarke CA, Lu Y, West DW, Bernstein L, Wang SS, Horn-Ross PL. Dietary phytocompounds and risk of lymphoid malignancies in the California Teachers Study cohort. Cancer Causes Control 2011;22:237–49.
31. La Vecchia C, Chatenoud L, Negri E, Franceschi S. Session: whole cereal grains, fibre and human cancer wholegrain cereals and cancer in Italy. Proc Nutr Soc 2003;62:45–9. 32. Lim U, Freedman DM, Hollis BW, Horst RL, Purdue MP, Chatterjee N, Weinstein SJ,
Morton LM, Schatzkin A, Virtamo J, Linet MS, Hartge P, et al. A prospective investigation of serum 25-hydroxyvitamin D and risk of lymphoid cancers. Int J cancer 2009;124:979– 86.
33. Negri E, La Vecchia C, Franceschi S, D’Avanzo B, Parazzini F. Vegetable and fruit consumption and cancer risk. Int J cancer 1991;48:350–4.
34. Tavani A, La Vecchia C, Gallus S, Lagiou P, Trichopoulos D, Levi F, Negri E. Red meat intake and cancer risk: a study in Italy. Int J cancer 2000;86:425–8.
35. Visioli F, Grande S, Bogani P, Galli C. The role of antioxidants in the mediterranean diets: focus on cancer. Eur J Cancer Prev 2004;13:337–43.
36. Thompson CA, Habermann TM, Wang AH, Vierkant RA, Folsom AR, Ross JA, Cerhan JR. Antioxidant intake from fruits, vegetables and other sources and risk of non-Hodgkin’s lymphoma: the Iowa Women’s Health Study. Int J cancer 2010;126:992–1003.
37. Holtan SG, O’Connor HM, Fredericksen ZS, Liebow M, Thompson CA, Macon WR, Micallef IN, Wang AH, Slager SL, Habermann TM, Call TG, Cerhan JR. Food-frequency questionnaire-based estimates of total antioxidant capacity and risk of non-Hodgkin lymphoma. Int J cancer 2012;131:1158–68.
38. Chiu Y-H, Bertrand KA, Zhang S, Laden F, Epstein MM, Rosner BA, Chiuve S, Campos H, Giovannucci EL, Chavarro JE, Birmann BM. A prospective analysis of circulating
saturated and monounsaturated fatty acids and risk of non-Hodgkin lymphoma. Int J
cancer 2018;
39. Smedby KE, Ponzoni M. The aetiology of B-cell lymphoid malignancies with a focus on chronic inflammation and infections. J Intern Med 2017;282:360–70.
40. Saberi Hosnijeh F, Krop EJM, Scoccianti C, Krogh V, Palli D, Panico S, Tumino R, Sacredote C, Nawroly N, Portengen L, Linseisen J, Vineis P, et al. Plasma cytokines and future risk of non-Hodgkin lymphoma (NHL): a case-control study nested in the Italian European Prospective Investigation into Cancer and Nutrition. Cancer Epidemiol
Biomarkers Prev 2010;19:1577–84.
41. Purdue MP, Lan Q, Bagni R, Hocking WG, Baris D, Reding DJ, Rothman N. Prediagnostic serum levels of cytokines and other immune markers and risk of non-hodgkin lymphoma.
Cancer Res 2011;71:4898–907.
42. Gu Y, Shore RE, Arslan AA, Koenig KL, Liu M, Ibrahim S, Lokshin AE, Zeleniuch-Jacquotte A. Circulating cytokines and risk of B-cell non-Hodgkin lymphoma: a prospective study. Cancer Causes Control 2010;21:1323–33.
43. Shivappa N, Steck SE, Hurley TG, Hussey JR, Hebert JR. Designing and developing a literature-derived, population-based dietary inflammatory index. Public Health Nutr 2014;17:1689–96.
44. Shivappa N, Hebert JR, Taborelli M, Montella M, Libra M, Zucchetto A, Crispo A, Grimaldi M, La Vecchia C, Serraino D, Polesel J. Dietary inflammatory index and non-Hodgkin lymphoma risk in an Italian case-control study. Cancer Causes Control 2017;28:791–9. 45. Sant M, Allemani C, Tereanu C, De Angelis R, Capocaccia R, Visser O, Marcos-Gragera
R, Maynadié M, Simonetti A, Lutz JM, Berrino F, Hackl M, et al. Incidence of hematologic malignancies in Europe by morphologic subtype: Results of the HAEMACARE project.
Blood 2010;116:3724–34.
46. Moreno LA, Sarria A, Popkin BM. The nutrition transition in Spain: a European Mediterranean country. Eur J Clin Nutr 2002;56:992–1003.
47. Slimani N, Kaaks R, Ferrari P, Casagrande C, Clavel-Chapelon F, Lotze G, Kroke A, Trichopoulos D, Trichopoulou A, Lauria C, Bellegotti M, Ocke MC, et al. European
Prospective Investigation into Cancer and Nutrition (EPIC) calibration study: rationale, design and population characteristics. Public Health Nutr 2002;5:1125–45.
48. Ferrari P, Kaaks R, Fahey MT, Slimani N, Day NE, Pera G, Boshuizen HC, Roddam A, Boeing H, Nagel G, Thiebaut A, Orfanos P, et al. Within- and between-cohort variation in measured macronutrient intakes, taking account of measurement errors, in the European Prospective Investigation into Cancer and Nutrition study. Am J Epidemiol 2004;160:814– 22.
49. Molina-Montes E, Sanchez M-J, Buckland G, Bueno-de-Mesquita HBA, Weiderpass E, Amiano P, Wark PA, Kuhn T, Katzke V, Huerta JM, Ardanaz E, Quiros JR, et al. Mediterranean diet and risk of pancreatic cancer in the European Prospective Investigation into Cancer and Nutrition cohort. Br J Cancer 2017;116:811–20.
Supplementary material
Table S1: Association between adherence to the arMED score and risk of lymphoma, HL, NHL and mature B-cell NHL by sex, age, alcohol consumption and smoking in the EPIC study.
Table S2: Association between adherence to the arMED score including alcohol and risk of lymphoma and its subtypes in the EPIC study.
Figure S1: Association between adherence to the arMED score and risk of lymphoma, Hodgkin lymphoma and non-Hodgkin lymphoma by country in the EPIC study.
Table S1
HR, hazard ratio; CI, confidence interval; HL, Hodgkin lymphoma; NHL, non-Hodgkin lymphoma; HL, Hodgkin lymphoma; DLBCL, diffuse large B-cell lymphoma; FL, follicular lymphoma; CLL/SLL, chronic lymphocytic leukemia/small lymphocytic leukemia; MM/PCN, multiple myeloma/ plasma cell neoplasm; Other B-cell (those cases for which the mature B-cell NHL subtype is unknown or does not fall within the more common subtypes)
1 NHL subtypes, excluding 37 precursor NHL and 37 individuals with NHL without B- or T-cell information.
2HR per 1-unit increase in the arMED; Cox proportional hazard model stratified by age (in 1-year categories), center and sex and further adjusted for body mass index, total energy intake, education, height, physical activity, smoking status, and alcohol intake.
3P-value of Cox proportional model fitted with the arMED continuous variable. 4P-value for interaction based upon the likelihood ratio (LR) test
5Not estimated due to small sample size (1 failure).
6Alcohol intake at recruitment categorized according to rMED categories for alcohol consumption: moderate (5–25 g/day for women and 10–50 g/day for men), low (below the sex-specific range) and high (above the sex-specific range).
In bold: p<0.05 Lymphoma (n=3,136) (n=135) HL NHL 1 (n=2,606) Mature B-cell (n=2,402) HR2 (95% CI) P-value3 HR2 (95% CI) P-value3 HR2 (95% CI) P-value3 HR2 (95% CI) P-value3 Age <50 1.00 (0.96; 1.04) 0.93 0.93 (0.83; 1.06) 0.28 1.01 (0.96; 1.05) 0.80 1.00 (0.96; 1.05) 0.85 50-75 0.98 (0.96; 0.99) 0.01 0.94 (0.84; 1.04) 0.21 0.98 (0.96; 1.00) 0.03 0.98 (0.96; 1.00) 0.05 >75 1.19 (0.98; 1.46) 0.09 Not estimated5 - 1.23 (0.97; 1.57) 0.09 1.16 (0.91; 1.48) 0.23 P-valueint4 0.79 0.21 0.53 0.59 Sex Men 0.98 (0.96; 1.01) 0.21 0.98 (0.86; 1.11) 0.72 0.98 (0.95; 1.01) 0.20 0.98 (0.95; 1.01) 0.15 Women 0.98 (0.96; 1.00) 0.08 0.90 (0.81; 1.00) 0.04 0.99 (0.96; 1.01) 0.27 0.99 (0.97; 1.01) 0.37 P-valueint4 0.75 0.61 0.99 0.80 Smoking status Never 0.98 (0.96; 1.00) 0.11 0.95 (0.84; 1.08) 0.42 0.98 (0.95; 1.01) 0.14 0.98 (0.95; 1.01) 0.12 Former 0.99 (0.96; 1.02) 0.54 1.01 (0.87; 1.17) 0.89 0.98 (0.95; 1.02) 0.36 0.98 (0.95; 1.02) 0.34 Current 0.98 (0.94; 1.01) 0.24 0.85 (0.74; 0.99) 0.03 1.00 (0.96; 1.05) 0.86 1.01 (0.97; 1.05) 0.74 P-valueint4 0.15 0.41 0.86 0.90
Alcohol intake (g/day)5
Low 0.98 (0.96; 1.01) 0.20 0.94 (0.84; 1.04) 0.23 0.99 (0.96; 1.01) 0.37 0.99 (0.97; 1.02) 0.51 Moderate 0.92 (0.86; 0.97) 0.004 0.94 (0.68; 1.28) 0.68 0.93 (0.87; 0.99) 0.02 0.92 (0.87; 0.99) 0.02 High 0.99 (0.97; 1.02) 0.51 0.93 (0.81; 1.06) 0.28 0.99 (0.96; 1.02) 0.52 0.99 (0.96; 1.02) 0.46
Table S2
arMED score Low
(0-7) Medium (6-10) (11-18) High P-value trend4 1-unit increase P-value5 Lymphoma, n 1,016 1,371 749 HR1 (95% CI) Ref 0.95 (0.86; 1.03) 0.89 (0.78; 1.02) 0.09 0.98 (0.97; 1.00) 0.04 HR2 (95% CI) Ref 0.95 (0.87; 1.04) 0.90 (0.79; 1.03) 0.13 0.99 (0.97; 1.00) 0.06 HL, n 39 65 31 HR1 (95% CI) Ref 0.84 (0.54; 1.30) 0.64 (0.34; 1.21) 0.17 0.93 (0.86; 0.99) 0.03 HR2 (95% CI) Ref 0.84 (0.54; 1.32) 0.64 (0.33; 1.22) 0.18 0.93 (0.86; 0.99) 0.04 NHL, n 835 1,128 643 HR1 (95% CI) Ref 0.95 (0.86; 1.04) 0.92 (0.79; 1.06) 0.20 0.99 (0.97; 1.00) 0.07 HR2 (95% CI) Ref 0.95 (0.86; 1.05) 0.93 (0.80; 1.07) 0.27 0.99 (0.97; 1.00) 0.10 NHL subtypes3 Mature T/NK-cell, n 47 54 29 HR1 (95% CI) Ref 0.88 (0.57; 1.35) 1.10 (0.58; 2.09) 0.97 1.03 (0.96; 1.10) 0.48 HR2 (95% CI) Ref 0.88 (0.57; 1.36) 1.14 (0.60; 2.19) 0.89 1.03 (0.96; 1.11) 0.43 Mature B-cell, n 766 1,043 593 HR1 (95% CI) Ref 0.95 (0.85; 1.05) 0.92 (0.79; 1.07) 0.23 0.98 (0.97; 1.00) 0.04 HR2 (95% CI) Ref 0.95 (0.86; 1.05) 0.93 (0.80; 1.09) 0.31 0.98 (0.97; 1.00) 0.06 DLBCL, n 147 245 96 HR1 (95% CI) Ref 1.02 (0.82; 1.27) 0.80 (0.56; 1.13) 0.35 0.96 (0.92; 0.99) 0.02 HR2 (95% CI) Ref 1.04 (0.83; 1.29) 0.81 (0.57; 1.16) 0.43 0.96 (0.92; 0.99) 0.02 FL, n 111 154 116 HR1 (95% CI) Ref 1.00 (0.78; 1.29) 1.25 (0.88; 1.78) 0.30 1.01 (0.97; 1.05) 0.62 HR2 (95% CI) Ref 1.00 (0.77; 1.30) 1.26 (0.88; 1.81) 0.28 1.01 (0.97; 1.05) 0.62 CLL/SLL, n 182 213 142 HR1 (95% CI) Ref 1.04 (0.84; 1.30) 1.00 (0.73; 1.38) 0.89 0.99 (0.96; 1.03) 0.78 HR2 (95% CI) Ref 1.03 (0.83; 1.28) 0.99 (0.71; 1.37) 0.98 0.99 (0.96; 1.03) 0.66 MM/PCN, n 226 283 167 HR1 (95% CI) Ref 0.81 (0.66; 0.98) 0.85 (0.64; 1.14) 0.14 0.99 (0.96; 1.02) 0.57 HR2 (95% CI) Ref 0.87 (0.65; 1.17) 0.87 (0.65; 1.17) 0.20 1.00 (0.96; 1.03) 0.82 Other B-cell ,n 100 148 72 HR1 (95% CI) Ref 0.90 (0.68; 1.18) 0.76 (0.50; 1.14) 0.19 0.96 (0.92; 1.00) 0.06 HR2 (95% CI) Ref 0.90 (0.68; 1.20) 0.76 (0.50; 1.16) 0.22 0.96 (0.91; 1.00) 0.06 HR, hazard ratio; CI, confidence interval; HL, Hodgkin lymphoma; NHL, non-Hodgkin lymphoma; HL, Hodgkin lymphoma; DLBCL, diffuse large B-cell lymphoma; FL, follicular lymphoma; CLL/SLL, chronic lymphocytic leukemia/small lymphocytic leukemia; MM/PCN, multiple myeloma/ plasma cell neoplasm; Other B-cell (those cases for which the mature B NHL subtype is unknown or does not fall within the more common subtypes).
3NHL subtypes, excluding 37 precursor NHL and 37 individuals with NHL without B- or T-cell information. 4P-value of Cox proportional model fitted with the arMED continuous variable.
5P-value for interaction based upon the likelihood ratio (LR) test In bold: P-value<0.05
Figure S1
HR, hazard ratio; CI, confidence interval; HL, Hodgkin lymphoma; NHL, non-Hodgkin lymphoma; HL, Hodgkin lymphoma; DLBCL, diffuse large B-cell lymphoma; FL, follicular lymphoma; CLL/SLL, chronic lymphocytic leukemia/small lymphocytic leukemia; MM/PCN, multiple myeloma/ plasma cell neoplasm; Other B-cell (those cases for which the
1HR per 1-unit increase in the arMED; Cox proportional hazard model stratified by age (in 1-year categories), center and sex and further adjusted for body mass index, total energy intake, education, height, physical activity, smoking status, and alcohol intake.
Table and figure legend
Table 1. Distribution of lymphoma cases in the EPIC study.
Table 2. Baseline characteristics of participants in the EPIC study according to adherence to the arMED score. Table 3. Association between adherence to the arMED score and risk of lymphoma and its subtypes in the EPIC study.
Table 1
Lymphoma subtypes NHL subtypes1 Mature B-cell subtypes
Total cohort Person-years Overall NHL HL NOS Mature B-cell Mature T /NK-cell DLBCL FL CLL/SLL MM/PCN Other B-cell mean (SD) arMED
Denmark 55,014 815,096.8 631 538 29 64 506 23 121 78 118 123 66 5.9 (2.4) France 67,403 869,362.5 228 216 11 1 205 8 40 44 44 45 32 8.5 (2.4) Germany 48,557 504,479.0 231 190 13 28 170 12 30 20 39 55 26 6.4 (2.1) Greece 26,048 281,283.6 62 44 3 15 38 2 3 3 13 15 4 11.8 (1.7) Italy 44,545 630,951.3 298 241 15 42 218 11 38 33 44 73 30 10.1 (2.1) Norway 33,975 452,171.1 163 147 5 11 129 14 26 31 26 24 22 7.7 (2.0) Spain 39,989 637,947.4 241 211 14 16 194 10 35 27 51 51 30 10.4 (2.2) Sweden 48,674 801,130.2 517 381 13 123 344 20 57 48 74 132 33 4.6 (2.1) The Netherlands 36,539 524,670.7 201 186 7 8 172 10 43 26 41 43 19 5.6 (2.1) United Kingdom 75,416 1,122,765 564 452 25 87 426 20 95 71 87 115 58 8.6 (2.5) Total 476,160 6,639,857.5 3,136 2,606 135 395 2,402 130 488 381 537 676 320 7.8 (3.0)
NHL, non-Hodgkin lymphoma; HL, Hodgkin lymphoma; NOS, not otherwise specified; DLBCL, diffuse large B-cell lymphoma; FL, follicular lymphoma; CLL/SLL, chronic lymphocytic leukemia/small lymphocytic leukemia; MM/PCN, multiple myeloma/ plasma cell neoplasm; Other B-cell (those cases for which the mature B-cell NHL subtype is unknown or does not fall within the more common subtypes); arMED, adapted relative Mediterranean diet; SD, standard deviation.
Table 2
arMED score
Total cohort Low (0-5)
(mean 3.9) Medium (6-9) (mean 7.5) High (10-16) (mean 11.4)
Total cohort, n 476,160 116,128 214,649 145,383
Sex (%)
Men 29.9 45.1 25.0 24.9
Women 70.1 54.9 75.0 75.1
Age at recruitment (mean [SD], years) 51.2 (9.9) 52.0 (10.0) 51.4 (9.5) 50.3 (10.4)
Energy intake (mean [SD], kcal/day) 2,075.1 (619.2) 2,178.5 (648.0) 2,039.8 (604.1) 2,044.6 (608.6)
Alcohol intake (median [25th-75th percentile], g/day) 5.3 (0.9; 14.9) 6.2 (1.4; 17.0) 5.6 (1.1; 15.2) 4.2 (0.4; 13.0)
BMI (mean [SD], kg/m2) 25.4 (4.3) 25.6 (4.1) 25.1 (4.1) 25.8 (4.5) Height (mean [SD], cm) 166.0 (8.9) 169.6 (9.2) 165.9 (8.5) 163.2 (8.4) Smoking status (%) Never 49.0 41.5 49.2 54.5 Former 26.6 27.7 27.7 24.3 Current 22.4 29.6 20.8 19.0 Unknown 2.0 1.24 2.3 2.3 Physical activity (%) Inactive 21.0 17.6 18.1 27.8 Moderately inactive 32.9 31.7 33.6 32.9 Moderately active 26.4 25.2 28.4 24.3 Active 17.9 22.3 17.8 14.5 Unknown 1.85 3.2 2.0 0.5 Educational level (%) None 4.4 0.5 2.4 10.5 Primary school 25.6 30.2 22.4 26.6 Technical/professional school 22.2 30.5 23.8 13.3 Secondary school 20.4 17.1 21.9 20.9 University 23.8 20.0 25.2 24.8 Unknown 3.6 1.7 4.3 3.9
Table 3
arMED score
Low Medium High P-valuetrend4 1-unit increase
P-value5 Lymphoma, n 1,016 1,371 749 HR1 (95% CI) Ref 0.92 (0.84; 1.01) 0.90 (0.79; 1.02) 0.08 0.98 (0.97; 1.00) 0.02 HR2 (95% CI) Ref 0.93 (0.85; 1.02) 0.91 (0.80; 1.03) 0.12 0.98 (0.97; 1.00) 0.03 HL, n 39 65 31 HR1 (95% CI) Ref 0.96 (0.61; 1.51) 0.64 (0.34; 1.18) 0.15 0.93 (0.86; 1.00) 0.07 HR2 (95% CI) Ref 0.97 (0.61; 1.53) 0.64 (0.34; 1.19) 0.16 0.93 (0.86; 1.01) 0.07 NHL, n 835 1,128 643 HR1 (95% CI) Ref 0.90 (0.81; 0.99) 0.93 (0.81; 1.06) 0.23 0.98 (0.97; 1.00) 0.06 HR2 (95% CI) Ref 0.90 (0.82; 1.00) 0.94 (0.82; 1.08) 0.31 0.98 (0.97; 1.00) 0.10 NHL subtypes3 Mature T/ NK-cell, n 47 54 29 HR1 (95% CI) Ref 0.73 (0.47; 1.13) 0.76 (0.42; 1.40) 0.31 0.99 (0.91; 1.07) 0.72 HR2 (95% CI) Ref 0.72 (0.47; 1.14) 0.78 (0.42; 1.44) 0.35 0.99 (0.91; 1.07) 0.80 Mature B-cell, n 766 1,043 593 HR1 (95% CI) Ref 0.91 (0.82; 1.01) 0.94 (0.82; 1.08) 0.31 0.98 (0.97; 1.00) 0.08 HR2 (95% CI) Ref 0.91 (0.82; 1.01) 0.95 (0.82; 1.10) 0.40 0.98 (0.97; 1.00) 0.11 DLBCL, n 147 245 96 HR1 (95% CI) Ref 1.09 (0.87; 1.37) 0.83 (0.60; 1.14) 0.35 0.96 (0.92; 1.00) 0.07 HR2 (95% CI) Ref 1.10 (0.88; 1.39) 0.84 (0.61; 1.17) 0.43 0.96 (0.93;1.01) 0.09 FL, n 111 154 116 HR1 (95% CI) Ref 0.84 (0.64; 1.11) 1.25 (0.90; 1.76) 0.19 1.02 (0.97; 1.07) 0.45 HR2 (95% CI) Ref 0.84 (0.64; 1.11) 1.27 (0.90; 1.79) 0.17 1.02 (0.97; 1.07) 0.43 CLL/SLL, n 182 213 142 HR1 (95% CI) Ref 0.80 (0.64; 0.99) 0.95 (0.70; 1.27) 0.54 0.99 (0.96; 1.03) 0.75 HR2 (95% CI) Ref 0.78 (0.63; 0.98) 0.92 (0.68; 1.25) 0.45 0.99 (0.95; 1.03) 0.64 MM/PCN, n 226 283 167 HR1 (95% CI) Ref 0.93 (0.77; 1.14) 0.98 (0.75; 1.28) 0.82 0.98 (0.95; 1.02) 0.40 HR2 (95% CI) Ref 0.95 (0.78; 1.16) 1.01 (0.77; 1.34) 0.98 0.99 (0.95; 1.03) 0.55 Other B-cell ,n 100 148 72 HR1 (95% CI) Ref 0.85 (0.64; 1.13) 0.67 (0.45; 1.00) 0.05 0.96 (0.91; 1.00) 0.08 HR2 (95% CI) Ref 0.85 (0.64; 1.14) 0.68 (0.46; 1.01) 0.06 0.96 (0.91; 1.00) 0.09 arMED: adapted relative Mediterranean dietary score; n, number of cases; HR, hazard ratio; CI, confidence interval; HL, Hodgkin lymphoma; NHL, non-Hodgkin lymphoma; HL, Hodgkin lymphoma; DLBCL, diffuse large B-cell lymphoma; FL, follicular lymphoma; CLL/SLL, chronic lymphocytic leukemia/small lymphocytic leukemia; MM/PCN, multiple myeloma/ plasma cell neoplasm; Other B-cell (those cases for which the mature B NHL subtype is unknown or does not fall within the more common subtypes); arMED, adapted relative Mediterranean diet; SD, standard deviation.
1Basic model: Cox proportional hazard model stratified by age (in 1-year categories), center and sex
2Multivariate model: Cox proportional hazard model stratified by age (in 1-year categories), center and sex, and further adjusted for body mass index, total energy intake, educational level, height, physical activity, smoking status, and alcohol intake.
3 NHL subtypes, excluding 37 precursor NHL and 37 individuals with NHL without B- or T-cell information. 4P-value of Cox proportional model fitted with the arMED ordinal variable as continuous to test for lineal trend. 5P-value of Cox proportional model fitted with the arMED continuous variable.
